Tones and Features Phonetic and Phonological Perspectives

382 Pages·2011·1.576 MB·English

by  John A. Goldsmith

Tones and Features
Studies in Generative Grammar 107
Editors
Harry van der
Hulst Jan Koster
Henk van Riemsdijk
De Gruyter Mouton
Tones and Features
Phonetic and Phonological Perspectives
edited by
John A. Goldsmith, Elizabeth Hume,
and W. Leo Wetzels
De Gruyter Mouton
The series Studies in Generative Grammar was formerly published by Foris Publications Holland.
ISBN 978-3-11-024621-6
e-ISBN 978-3-11-024622-3
ISSN 0167-4331
Library of Congress Cataloging-in-Publication Data
Tones and features : phonetic and phonological perspectives / edited by John A.
Goldsmith, Elizabeth Hume, Leo Wetzels.
p. cm. — (Studies in generative grammar; 107)
Includes bibliographical references and index.
ISBN 978-3-11-024621-6 (alk. paper)
1. Phonetics. 2. Grammar, Comparative and general—Phonology.
I. Goldsmith, John A., 1951- II. Hume, Elizabeth V., 1956- III. Wetzels, Leo.
P217.T66 2011
414'.8—dc23
2011030930
Bibliographic information published by the Deutsche Nationalbibliothek
The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliogra(cid:191) e;
detailed bibliographic data are available in the Internet at http://dnb.d-nb.de.
© 2010 Walter de Gruyter GmbH & Co. KG, Berlin/Boston
Typesetting: Re(cid:191) neCatch Ltd, Bungay, Suffolk
Printing: Hubert & Co. GmbH & Co. KG, Göttingen
(cid:2) Printed on acid-free paper
Printed in Germany
www.degruyter.com
Contents
Preface vii
John Goldsmith, Elizabeth Hume and W. Leo Wetzels
1. The representation and nature of tone
Do we need tone features? 3
G. N. Clements, Alexis Michaud and Cédric Patin
Rhythm, quantity and tone in the Kinyarwanda verb 25
John Goldsmith and Fidèle Mpiranya
Do tones have features? 50
Larry M. Hyman
Features impinging on tone 81
David Odden
Downstep and linguistic scaling in Dagara-Wulé 108
Annie Rialland and Penou-Achille Somé
2. The representation and nature of phonological features
Crossing the quantal boundaries of features: Subglottal
resonances and Swabian diphthongs 137
Grzegorz Dogil, Steven M. Lulich,
Andreas Madsack, and Wolfgang Wokurek
Voice assimilation in French obstruents: Categorical or gradient? 149
Pierre A. Hallé and Martine Adda-Decker
An acoustic study of the Korean fricatives /s, s'/:
implications for the features [spread glottis] and [tense] 176
Hyunsoon Kim and Chae-Lim Park
vi Contents
Autosegmental spreading in Optimality Theory 195
John J. McCarthy
Evaluating the effectiveness of Uni(cid:191) ed Feature Theory
and three other feature systems. 223
Jeff Mielke, Lyra Magloughlin, and Elizabeth Hume
Language-independent bases of distinctive features 264
Rachid Ridouane, G. N. Clements and Rajesh Khatiwada
Representation of complex segments in Bulgarian 292
Jerzy Rubach
Proposals for a representation of sounds based on their
main acoustico-perceptual properties 306
Jacqueline Vaissière
The representation of vowel features and vowel
neutralization in Brazilian Portuguese (southern dialects) 331
W. Leo Wetzels
Index 361
Preface
The papers in this volume are all concerned with two current topics in
phonology: the treatment of features, and the treatment of tone. Most of them
grew out of a conference at the University of Chicago’s Paris Center in June
of 2009 which was organized by friends and colleagues of Nick Clements in
tribute to decades of contributions that he had made to the (cid:191) eld of phonology,
both in the United States and in France. Nick’s work served as a natural focus
for the discussions and interactions that resulted in the papers that the reader
will (cid:191) nd in this book. We, the editors, would like to say a bit about Nick’s
career and his work in order to set the context.
1. G. N. Clements
Nick was an undergraduate at Yale University, and received his PhD from
the School of Oriental and African Studies, University of London, for a
dissertation on the verbal syntax of Ewe in 1973, based on work that he did in
the (cid:191) eld. In the 1970s, he spent time as a post-doctoral scholar at MIT and then
as a faculty member in the Department of Linguistics at Harvard University.
Throughout this period he published a series of very in(cid:192) uential articles and
books on areas in phonological theory, a large portion of which involved
linguistic problems arising out of the study of African languages. His work
in this period played an essential role in the development of autosegmental
phonology, and his work in the 1980s, when he was a professor of linguistics
at Cornell University, was crucial in the development of many of the current
views on features, feature geometry, sonority, and syllabi(cid:191) cation. He worked
closely with students throughout this time—including one of us, Elizabeth
Hume—at Cornell. He also co-wrote books with several phonologists
(Morris Halle, Jay Keyser, John Goldsmith) and collaborated on many
research projects.
In 1991, Nick moved to Paris, where he and his wife, Annie Rialland,
worked together on projects in phonetics, phonology, and many other things,
both linguistic and not. Visiting Nick in Paris became an important thing for
phonologists to do when they had the opportunity to come to Paris. Over the
next twenty years or so Nick continued to work sel(cid:192) essly and generously
viii Preface
with students and more junior scholars, and was widely sought as an invited
speaker at conferences.
Nick passed away a few months after the conference, late in the summer
of 2009. Many of his friends (and admirers) in the discipline of phonology
had been able to express their admiration for his contributions through their
papers and their kind words at the time of the conference in June. This book is
offered as a more permanent but equally heartfelt statement of our affection
and respect for Nick’s work in phonology and in linguistics more broadly.
2. Tone
The proper treatment of tonal systems has long been an area of great activity
and curiosity for phonologists, and for several reasons. Tonal systems appear
exotic at (cid:191) rst blush to Western European linguists, and yet are common
among languages of the world. The phonology of tone is rich and complex, in
ways that other subdomains of phonology do not illustrate, and yet each step
in our understanding of tonal systems has shed revelatory light on the proper
treatment of other phonological systems. At every turn, tonal systems stretch
our understanding of fundamental linguistic concepts: many languages
exhibit tonal contrasts, in the sense that there are lexical contrasts that are
physically realized as different patterns of fundamental frequency distributed
globally over a word. But from a phonological point of view, words are not
unanalyzable: far from it—they are composed in an organized fashion from
smaller pieces, some mixture of feet, syllables, and segments. Breaking a
pitch pattern (when considering an entire word) into pieces that are logically
related to phonological or morphological subpieces (which is ultimately
ninety percent of a phonologist’s synchronic responsibility) has proven time
and time again to be an enormous challenge in the arena of tone. One of
the classic examples of this challenge can be found in Clements and Ford’s
paper (1979) on Kikuyu tone. In Kikuyu, the surface tone of each syllable
is essentially the expression of the previous syllable’s tonal speci(cid:191) cation.
Each syllable (often, though not always, a distinct morpheme) thus has an
underlying – we are tempted to say, a logical—tone speci(cid:191) cation, but that
speci(cid:191) cation is realized just slightly later in the word than the syllable that
comprises the other part of the underlying form. Morphemes in such a system
show utter disregard for any tendency to try to be realized in a uniform way
across all occurrences; tones seem to assert their autonomy and the privileges
that come with that, and use it to produce a sort of constant syncopation in
the beat of syllable against tone.
Preface ix
Is tone, then, different from other phonological features? This question
is directly posed by three papers in this volume, that by Nick Clements and
colleagues, that by Larry Hyman, and that by David Odden. Each is written
with the rich background of several decades of research on languages – largely
African tone languages, at least as far as primary research is concerned, but also
including the fruits of research done on Asian languages over decades as well. In
the end, Clements, Michaud, and Patin conclude that tonal features may well be
motivated in our studies of tonal systems, but the type of motivation is different
in kind from that which is familiar from the study of other aspects of phonology.
Hyman, for his part, is of a similar conviction: if tones are analyzed featurally
in the ultimate model of phonology, it is not a step towards discovering ultimate
similarity between tone and every other phonological thing: tone’s diversity
in its range of behavior keeps it distinct from other parts of phonology. David
Odden’s chapter also focuses on the motivation for tonal features. However, his
focus is on the types of evidence used to motivate a given feature. Along these
lines, he argues that tonal features, like other phonological features, are learned
on the basis of phonological patterning rather than on the basis of the physical
properties of the sounds (for related discussion, see Mielke 2008).
Goldsmith and Mpiranya’s contribution addresses not features for tone,
but rather one particular characteristic of tone that keeps it distinct from other
aspects of phonology: tone’s tendency to shift its point of realization (among a
word’s syllables) based on a global metrical structure which is erected on the
entire word. This is similar to the pattern we alluded to just above in Kikuyu,
but in Kinyarwanda, certain High tones shift their autosegmental association
in order to appear in weak or strong rhythmic positions: a bit of evidence that
rhythmicity is an important organization principle of tonal assignment, in at
least some languages, much like that seen in accent assignment and rarely, if
ever, seen in other aspects of a phonological system.
The theme of rhythmicity is continued in the paper by Annie Rialland and
Penou-Achille Somé. They hypothesize that there is a relationship between the
linguistic scaling in Dagara-Wulé, as manifested in downstep sequences, and the
musical scaling in the same culture, as found in an eighteen key xylophone. They
suggest that downstep scaling and xylophone scaling may share the property of
being comprised of relatively equal steps, de(cid:191) ned in terms of semitones.
3. Features
The hypothesis that the speech chain can be analyzed as a sequence of
discrete segments or phonemes, themselves decomposable into a set of